Learning SourceTarget Surface Patterns for Web-based Terminology Translation

This paper introduces a method for learn-ing to find translation of a given source term on the Web. In the approach, the source term is used as a query and part of patterns to retrieve and extract transla-tions in Web pages. The method involves using a bilingual term list to learn source-target surface patterns. At runtime, the given term is submitted to a search engine then the candidate translations are ex-tracted from the returned summaries and subsequently ranked based on the surface patterns, occurrence counts, and translit-eration knowledge. We present a proto-type called TermMine that applies the method to translate terms. Evaluation on a set of encyclopedia terms shows that the method significantly outperforms the state-of-the-art online machine translation systems.